the curve of Case 1 (Blue) is above the one of Case 3 (Red) if x > 4
For example, when x = 39, i.e., the vector of !happy = [1, 1, ..., 1]
case
happy
!happy
Case 1
[ 1 ]
[ 1, 1, ..., 1 ]
Case 3
[ 10 ]
[ 1, 1, ..., 1 ]
Intuitively, 10 is more significant than 1, i.e., the possibility of classifying to happy in Case 3 is higher than Case 1. And it is obvious that we got the opposite results lol
In the results of Method 3,
the curve of Case 1 (Blue) is above the one of Case 3 (Red) if x > 4
For example, when x = 39, i.e., the vector of
!happy
= [1, 1, ..., 1]Intuitively, 10 is more significant than 1, i.e., the possibility of classifying to happy in
Case 3
is higher thanCase 1
. And it is obvious that we got the opposite results lol